Control of Surface Tension at Liquid–Liquid Interfaces Using

Dec 13, 2011 - 2D constructs using these emulsion platforms provides a strategy for the ..... S. S.; Miranda, O. R.; Mo, Z.-H.; Rotello, V. M. Angew. ...
0 downloads 0 Views 332KB Size
Article pubs.acs.org/Langmuir

Control of Surface Tension at Liquid−Liquid Interfaces Using Nanoparticles and Nanoparticle−Protein Complexes Subinoy Rana,† Xi Yu,†,§ Debabrata Patra,†,§ Daniel F. Moyano,† Oscar R. Miranda,† Irshad Hussain,‡ and Vincent M. Rotello*,† †

Department of Chemistry, University of Massachusetts at Amherst, 710 North Pleasant Street, Amherst, Massachusetts 01003, United States ‡ Department of Chemistry, LUMS School of Science & Engineering (SSE), D.H.A., Lahore Cantt 54792, Pakistan S Supporting Information *

ABSTRACT: Subtle changes in the monolayer structure of nanoparticles (NPs) influence the interfacial behavior of both NPs and NP−protein conjugates. In this study, we use a series of monolayer-protected gold NPs to explore the role of particle hydrophobicity on their dynamic behavior at the toluene−water interface. Using dynamic surface tension measurements, we observed a linear decrease in the mesoequilibrium surface tension (γ) and faster dynamics as the hydrophobicity of the ligands increases. Further modulation of γ is observed for the corresponding NP− protein complexes at the charge-neutralization point.

1. INTRODUCTION Emulsions stabilized by nanoparticlesnovel surface-active materials, and emulsion stabilizers1have applications in diverse areas including catalysis,2,3 drug delivery,4 and food science.5 In these systems, nanoparticles (NPs) stabilize the liquid−liquid interface through the minimization of the interfacial energy of the system.6 In addition to the stabilization of the interface, the dynamic nature of the NP assembly at liquid−liquid interfaces can be used to generate defect-free organized nanostructures.7 The assembly of nanocrystals into 2D constructs using these emulsion platforms provides a strategy for the creation of novel multiscale materials with unique electronic, magnetic, and optical properties.8 Recently, the assembly of bionanoparticles such as proteins, virus particles, and other supramolecular protein assemblies has shown promise for creating highly organized biomolecular systems utilizing their monodisperse sizes, biocompatibility, and ease of functionalization at interfaces.9 Interparticle interaction at liquid−liquid interfaces is one of the key determinants in emulsion stabilization.10 Prior approaches to the creation of nanoparticle-stabilized emulsions have focused on the chemical cross-linking of NPs assembled at the oil−water interface.11 Recently, supramolecular interactions have been used to fabricate robust microcapsules.12 In particular, the charge-mediated assembly of NPs and proteins has been used to provide lateral stabilization of the emulsions.3 The key requirements of this approach are the amphiphilic nature of the resulting supramolecular assemblies and their instantaneous migration to the interface; the synergistic interaction of the NP−protein complex at the interface and the dynamics of interfacial adsorption are crucial factors governing the stability of liquid−liquid emulsions.13 Surface tension is a fundamental quantity that relates to the properties of the assembly and segregation dynamics of surface© 2011 American Chemical Society

active materials at phase interfaces and plays a crucial role in the process of emulsion formation and stabilization.14,15 In this article, we report the dynamic interfacial behavior of monolayer-protected gold NPs and NP−protein conjugates at the toluene−water interface using dynamic surface tension measurements that suggest a kinetic diffusion mechanism of interfacial assembly. These studies also indicate that increasing the hydrophobicity of the NP monolayer lowers the interfacial tension in a linear fashion, both with the particle alone and with the particle−protein assemblies. This investigation demonstrates the key role of the ligand hydrophobicity in the interfacial assembly of the particles and particle−protein complexes. These kinetics studies provide insight into the mechanism of interfacial assembly, providing guidelines for designing appropriate nanoparticles with which to prepare stable emulsions.

2. EXPERIMENTAL SECTION 2.1. Materials. All materials were purchased from Sigma-Aldrich and used without further purification. For all measurements, NPs and protein were dispersed in 5 mM phosphate-buffered saline (PBS) at pH 7.4. 2.2. Nanoparticle Synthesis. Nanoparticles were synthesized according to literature-reported procedures.16 In short, 1-pentanethiolcoated gold NPs (C5-NP) were fabricated using the NaBH4 reduction of HAuCl4− salt. Ligands with varying hydrophobic chains were then synthesized and place-exchanged onto 2 nm core C5-NPs. All five NPs in the study were prepared from the same C5-NP core so that the NPs differed only in terms of the desired surface functionalities. Special Issue: Bioinspired Assemblies and Interfaces Received: October 13, 2011 Revised: December 9, 2011 Published: December 13, 2011 2023

dx.doi.org/10.1021/la204017z | Langmuir 2012, 28, 2023−2027

Langmuir

Article

2.3. Tensiometer Measurements. The dynamic surface tension of the NPs and NP−protein conjugates at the toluene−water interface was measured using the pendant drop method (OCA20, Dataphysics, Stuttgart). The setup of the pendant drop apparatus is shown elsewhere.17 A syringe filled with a solution of NPs or NP−protein complexes connected to a needle was fixed vertically with the needle immersed in the toluene phase. A small amount of solution was injected from the syringe to form a drop. The variation of drop shape with time was captured by automated camera at particular time intervals, and the interfacial tension (γ) was estimated by data fitting using the Laplace−Young equation

γ=

Δρgde H

(1)

where Δρ is the density difference between the liquid drop and its surrounding medium, g is the gravitational acceleration, de is the largest horizontal diameter of the drop, and H is a function of Sn (= dn/de), in which dn is the horizontal diameter at a distance equal to de (n/10) from the bottom of the drop. All experiments were performed at room temperature (23 ± 1 °C).

2.4. DLS and Zeta Potential Measurements. The hydrodynamic radius and zeta potential of NPs and NP−protein complexes were measured by dynamic light scattering (DLS) in 5 mM phosphate buffer (pH 7.4) using a Malvern ZetasizerNano ZS instrument. All of the NP−protein complexes were incubated for 30 min prior to measurements. Solutions of NPs and NP−protein complexes in different ratios were placed in a cuvette, and the average of three measurements was considered. 2.5. Curve Fitting and Partition Coefficient Calculation. Both linear and nonlinear curve fitting were performed using Origin 7 software following an iterative method. The log P values were estimated for the headgroup of each ligand (starting from the quaternary nitrogen, e.g., (CH3)2NCH3 for NP1, (CH3)2NCH2CH3 for NP2, etc.) using molecular mechanics. The force field employed was an MMFF under TNCG methodology using octanol and water as the first and second solvents, respectively. The calculations were continued until gradient convergence occurred, using MacroModel 8.0 under the Maestro 9.5 interface.

Figure 1. (a) Structure of the NPs and their monolayer ligand shells. (b) Schematic of the assembly of particles at the oil−water interface. Initially the surface coverage is very low, and then the particles rapidly cover the interface. In the steady state, complete monolayer coverage is observed when packing rearrangement and conformational changes occur. (c) Schematic of NP−protein complex formation.

3. RESULTS AND DISCUSSION We synthesized five structurally related cationic gold NPs (Figure 1a) featuring 2 nm cores to isolate the role of NP surface hydrophobicity in interfacial assembly. These gold NPs provide multivalent binding with anionic proteins, forming complexes through electrostatic and hydrophobic interactions.18 Additionally, these bioinspired NPs feature a tetra(ethylene glycol) spacer in the ligand shell both to isolate the effects of the headgroup19 and to minimize the denaturation of the bound protein.20 We used acid phosphatase (PhosA) as a model anionic protein because of its relatively large size (Mw = 110 kDa, 12.6 nm × 20.7 nm × 7.3 nm) and negative charge at neutral pH (pI = 5.2). We first studied time-dependent pendant drop tensiometer measurements at the water−toluene interface for the particles (NP1−NP5) at 1 μM in 5 mM PBS buffer at pH 7.4. Figure 2 presents the dynamic surface tension (γt) plots for NP1−NP5, where the interfacial tension decreases with time, approaching an equilibrium value. In the early stage, the interfacial tension drops rapidly because of instantaneous self-assembly of the NPs at the interface. Once the droplet is mostly covered by NPs, the decrease in interfacial tension reaches a dynamic equilibrium where the rate of adsorption of NPs at the interface equals the rate of desorption (Figure 1b).17 It can be observed that the experimental meso-equilibrium γ (γeq) value decreases significantly with increasing hydrophobicity of the particles (Figure 2). NP4 and NP5, having ligand headgroups of same alkyl chain length and hence similar hydrophobicity, attain

Figure 2. Time dependence of the interfacial tension for gold NPs with different monolayer structures at the same bulk concentration of 1 μM at the toluene−water interface. The number near each plot indicates the γeq value at 10 000 s.

similar γeq values but exhibit different kinetic behaviors (vide infra). The variation in interfacial tension with different NPs can be explained by the effect of their surface functionality on the assembly at the oil−water interface. Hydrophobic NP4 and NP5 behave as amphiphilic surfactants at the water−toluene interface, resulting in the maximum reduction in the equilibrium interfacial tension. More hydrophilic NP2 and NP3 behave as moderately weak amphiphiles and intermediate 2024

dx.doi.org/10.1021/la204017z | Langmuir 2012, 28, 2023−2027

Langmuir

Article

γeq values are observed, and hydrophilic NP1 provides the smallest amount of effective interfacial tension reduction. In addition to providing enhanced surface activity, the increasing hydrophobicity of the nanoparticle accelerates the emulsification process. The kinetic behavior of the surface tension of the NPstabilized water-in-oil emulsions was obtained using γt versus time plots fitted to the empirical Hua and Rosen equation21 that has been widely employed for interpreting dynamic surface tension behaviors at interfaces. One form of the equation is given by eq 2

γt = γm +

γ0 − γm 1 + (t /t *)n

(2) Figure 3. Plot of γm vs the partition coefficient of the ligand headgroups on the particle monolayer.

where γt is the surface tension at any time t, γ0 is the surface tension of the pure solvent, γm is the surface tension at mesoequilibrium, t* is the half-time in reaching γm, and n is a dimensionless exponent. Assuming that the value of γ0 is held constant at the value of the pure buffer surface tension, there are three adjustable parameters: γm, t*, and n. These parameters were estimated by computer fitting of the measured dynamic surface tension data. By differentiating eq 2 with respect to t and substituting t for t*, the maximum surface tension decay rate, νmax, of γ is obtained as follows:

νmax =

⎛ dγ ⎞ n(γ0 − γm) = − ⎜ t⎟ 4t * ⎝ dt ⎠t *

(Supporting Information Table S1), the diffusion-controlled mechanism alone cannot explain the observed trend in the dynamics. Presumably, it is the adsorption rate at the interface that determines the kinetic parameters of the system. NPs with increased lipophilicity tend to possess higher adsorption rates toward the interface, and hence faster dynamics is expected. In fact, an increasing decay rate is observed for NP1 to NP5. However, NP5 exhibit a 2.3-fold faster decay rate of γt than NP4, even though their hydrophobicities are quite similar. Structural attributes such as the constrained cyclic conformation of NP5 compared to that of the flexible open chain of NP4 possibly help in attaining a higher decay rate for NP5 relative to that for NP4. The values of νmax and t* reflect the major role of intermolecular interactions at the liquid−liquid interface17,23 that influences the adsorption energy barrier induced by the nanoparticle film at the interface. Although t* decreases along the series of NPs, a drastic change is observed for NP3−NP5 compared to that for NP1 and NP2. It suggests that the increase in the alkyl chain length poses an adsorption energy barrier that leads to the faster attainment of meso-equilibrium. Again, the time required to reach equilibrium is much higher for NP4 than for NP5, reflecting a higher energy barrier for the former particle. This is further supported by the parameter n, which is related to the difference between the adsorption and desorption energies of the surface-active moieties.24 Because the decrease in the n value suggests a decrease in the adsorption rate relative to that of desorption, a higher adsorption barrier is expected by the adsorbed molecules for NP4 with an open hydrocarbon chain compared to NP5 having a cyclic chain. Taken together, the interfacial behaviors of the monolayerprotected NPs indicate the presence of an interaction-based mixed kinetic diffusion mechanism. We next explored the role of NP hydrophobicity in the stabilization of liquid−liquid interfaces using protein−particle assemblies. We prepared the NP−protein complexes by mixing positively charged NPs and negatively charged PhosA in PBS buffer, followed by incubation to facilitate complexation. From our previous studies, we found that NP−protein conjugates with reduced charge successfully stabilized the oil−water interface.3,4 To obtain neutral conjugates, we performed a zeta potential (ζ) titration of NPs with varying PhosA concentration (Figure 4a). Depending on the functional headgroups, each NP exhibits drastically different binding stoichiometry with PhosA (Figure 4b). At the obtained ratios, different NP−protein assembly structures could be envisaged25

(3)

All of the γt versus time plots were fitted nicely, with correlation coefficients being ≥0.99 (Supporting Information Figure S1), and the results are presented in Table 1. It can be Table 1. Dynamic Surface Tension Parameters Calculated from γt vs Time Plots at the Toluene−Water Interface NPs

γm (mN/m)

t* (s)

n

NP1 NP2 NP3 NP4 NP5

17.7 14.1 12.35 10.18 10.96

1100 907 158 102 53

0.807 0.596 0.442 0.517 0.641

νmax (mN/m s) 2.7 3.0 14.1 28.2 64.5

× × × × ×

10−3 10−3 10−3 10−3 10−3

observed that γm decreases from NP1 to NP4 as the hydrophobicity increases. Also, NP4 and NP5 with similar hydrophobicity have similar γm values. We next investigated the quantitative relationship between the hydrophobicity of the ligand headgroup and the surface tension at the mesoequilibrium. We determined the computed octanol−water partition coefficient (Poct) of the NP headgroups, which can be used as a measure of hydrophobicity.22 We observed a linear decrease in γm with increasing hydrophobicity of the ligand headgroups as shown in Figure 3. The data set was well fit by eq 4 with a correlation of 0.965:

γm = 18.66 − 2.204 log10 Poct

(4)

The effect of the NP monolayer on their dynamics at the toluene−water interface, isolated from Hua−Rosen parameters t* and νmax, provides mechanistic insight into particle segregation at the interface. It can be observed that with increasing hydrophobicity from NP1 to NP5 the value of t* decreases gradually while νmax increases. Because the hydrodynamic size of the NPs increases from NP1 to NP5 2025

dx.doi.org/10.1021/la204017z | Langmuir 2012, 28, 2023−2027

Langmuir

Article

equilibrium surface tension values of NP alone and NP−protein are nearly constant regardless of the particle, suggesting that the interfacial energy can be predictably tuned with these systems.

4. CONCLUSIONS We have studied the behaviors of NPs and NP−protein complexes at a liquid−liquid interface. These studies demonstrate that subtle changes in the monolayer structure of the NPs play an important role in stabilizing the liquid− liquid interface. As the hydrophobicity of the ligand in the monolayer increases, NPs tend to create a more stable interface with an observed linear decrease in the meso-equilibrium surface tension. Notably, the dynamic interfacial behaviors of the NPs are governed by hydrophobicity and the structure of the NP layer at the interface, which set the energy barrier for their adsorption. It is clear that the dynamics of the assembly of the monolayer-protected NPs follows an interaction-based mixed kinetic diffusion mechanism where the adsorption of NPs at the interface is the rate-controlling step rather than diffusion. The kinetics studies also demonstrate that further modulation occurs after the complexation of the NPs with an electrostatically complementary protein. Given the important role that the interfacial assembly of particles plays in the formation and stability of a droplet in the emulsion process,15 these studies provide useful guidelines for designing appropriate NPs for emulsion and interfacial stabilization.

Figure 4. (a) Representative ζ-potential measurement data of the PhosA−NP4 conjugate prepared by varying the molar ratio of NPs to PhosA. The red line represents the best fit to the sigmoidal equation. (b) NP−protein ratios at ζ ≈ 0.



(Figure 1c) as contributing to the interfacial tension modulation. The optimal charge neutralization ratio was used for further interfacial tension measurements. The interfacial behavior of NP−PhosA complexes is shown in Figure 5a. It is evident that all five NP−PhosA complexes have lower interfacial tension values at equilibrium (quasi) and faster dynamics than do NPs alone. The NP1−PhosA complexes show the highest interfacial tension (γeq ≈ 13.6 mN/m) whereas NP5−PhosA complexes show the minimum interfacial tension (γeq ≈ 7.9 mN/m), indicating maximum stabilization at the oil−water interface. However, the Hua− Rosen model fails for the NP−PhosA complexes, and the data could not be fitted, presumably because of other processes such as particle−protein assembly. Significantly, the ratio of the

ASSOCIATED CONTENT

S Supporting Information *

Best-fit curves of the γt vs time data for NP1−NP5. Hydrodynamic diameter data obtained from DLS measurements. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (+1) 413-545-2058. Fax: (+1) 413-5452058. Author Contributions §

These authors contributed equally to this work.

Figure 5. (a) Interfacial tension vs time plot for NP−protein conjugates at the charge neutralization ratio. (b) Comparison of the surface tension values of the NPs and the corresponding NP−PhosA conjugates at 8000 s and their ratios. 2026

dx.doi.org/10.1021/la204017z | Langmuir 2012, 28, 2023−2027

Langmuir



Article

(b) Cardenas-valera, A. E.; Bailey, A. I. Colloids Surf., A 1995, 97, 1−12. (c) Wollenweber, C.; Makievski, A. V.; Miller, R.; Daniels, R. Colloids Surf., A 2000, 172, 91−101. (16) You, C.-C.; Miranda, O. R.; Gider, B.; Ghosh, P. S.; Kim, I.-B.; Erdogan, B.; Krovi, S. A.; Bunz, U. H. F.; Rotello, V. M. Nat. Nanotechnol. 2007, 2, 318−323. (17) Kutuzov, S.; He, J.; Tangirala, R.; Emrick, T.; Russell, T. P.; Boker, A. Phys. Chem. Chem. Phys. 2007, 9, 6351−6358. (18) De, M.; You, C.-C.; Srivastava, S.; Rotello, V. M. J. Am. Chem. Soc. 2007, 129, 10747−10753. (19) Moyano, D. F.; Rotello, V. M. Langmuir 2011, 27, 10376− 10385. (20) (a) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103−1169. (b) Nikolic, M. S.; Krack, M.; Aleksandrovic, V.; Kornowski, A.; Foerster, S.; Weller, H. Angew. Chem., Int. Ed. 2006, 45, 6577−6580. (21) Hua, X. Y.; Rosen, M. J. J. Colloid Interface Sci. 1988, 124, 652− 659. (22) Ghose, A. K.; Crippen, G. M. J. Comput. Chem. 1986, 7, 565− 577. (23) Du, K.; Glogowski, E.; Emrick, T.; Russell, T. P.; Dinsmore, A. D. Langmuir 2010, 26, 12518−12522. (24) Gao, T.; Rosen, M. J. J. Colloid Interface Sci. 1995, 172, 242− 248. (25) De, M.; Miranda, O. R.; Rana, S.; Rotello, V. M. Chem. Commun. 2009, 2157−2159.

ACKNOWLEDGMENTS Support from the Department of Energy (DE-FG0204ER46141) is acknowledged by V.M.R., and I.H. thanks Lahore University of Management Sciences (LUMS) for the start-up grant to initiate nanomaterials research at SSE. We thank Dr. Thomas P. Russell for providing access to the tensiometer.



REFERENCES

(1) Binks, B. P. Phys. Chem. Chem. Phys. 2007, 9, 6298−6299. (2) Crossley, S.; Faria, J.; Shen, M.; Resasco, D. E. Science 2010, 327, 68−72. (3) Samanta, B.; Yang, X.-C.; Ofir, Y.; Park, M.-H.; Patra, D.; Agasti, S. S.; Miranda, O. R.; Mo, Z.-H.; Rotello, V. M. Angew. Chem., Int. Ed. 2009, 48, 5341−5344. (4) Yang, X.-C.; Samanta, B.; Agasti, S. S.; Jeong, Y.; Zhu, Z.-J.; Rana, S.; Miranda, O. R.; Rotello, V. M. Angew. Chem., Int. Ed. 2011, 50, 477−481. (5) Dickinson, E. Curr. Opin. Colloid Interface Sci. 2010, 15, 40−49. (6) (a) Pieranski, P. Phys. Rev. Lett. 1980, 45, 569−572. (b) Binks, B. P.; Lumsdon, S. O. Langmuir 2000, 16, 8622−8631. (c) Dinsmore, A. D.; Hsu, M. F.; Nikolaides, M. G.; Marquez, M.; Bausch, A. R.; Weitz, D. A. Science 2002, 298, 1006−1009. (7) (a) Boeker, A.; He, J.; Emrick, T.; Russell, T. P. Soft Matter 2007, 3, 1231−1248. (b) Patra, D.; Sanyal, A.; Rotello, V. M. Chem. Asian J. 2010, 5, 2442−2453. (c) Binks, B. P.; Horozov, T. S. In Colloidal Particles at Liquid Interfaces; Cambridge University Press: Cambridge, U.K., 2006. (d) Kim, J.-W.; Lee, D.; Shum, H. C.; Weitz, D. A. Adv. Mater. 2008, 20, 3239−3243. (e) Binks, B. P.; Clint, J. H.; Fletcher, P. D. I.; Lees, T. J. G.; Taylor, P. Langmuir 2006, 22, 4100−4103. (8) (a) Lin, Y.; Skaff, H.; Boker, A.; Dinsmore, A. D.; Emrick, T.; Russell, T. P. J. Am. Chem. Soc. 2003, 125, 12690−12691. (b) Patra, D.; Malvankar, N.; Chin, E.; Tuominen, M.; Gu, Z.; Rotello, V. M. Small 2010, 6, 1402−1405. (c) Duan, H. W.; Wang, D. Y.; Sobal, N. S.; Giersig, M.; Kurth, D. G.; Mohwald, H. Nano Lett. 2005, 5, 949− 952. (9) (a) Lee, L. A.; Wang, Q. Nanomedicine 2006, 2, 137−49. (b) Fujii, S.; Aichi, A.; Muraoka, M.; Kishimoto, N.; Iwahori, K.; Nakamura, Y.; Yamashita, I. J. Colloid Interface Sci. 2009, 338, 222− 228. (c) He, J.; Niu, Z.; Tangirala, R.; Wan, J.-Y.; Wei, X.; Kaur, G.; Wang, Q.; Jutz, G.; Boeker, A.; Lee, B.; Pingali, S. V.; Thiyagarajan, P.; Emrick, T.; Russell, T. P. Langmuir 2009, 25, 4979−4987. (d) Kaur, G.; He, J.; Xu, J.; Pingali, S. V.; Jutz, G.; Boeker, A.; Niu, Z.; Li, T.; Rawlinson, D.; Emrick, T.; Lee, B.; Thiyagarajan, P.; Russell, T. P.; Wang, Q. Langmuir 2009, 25, 5168−5176. (e) Hermanson, K. D.; Huemmerich, D.; Scheibel, T.; Bausch, A. R. Adv. Mater. 2007, 19, 1810−1815. (10) Bresme, F.; Oettel, M. J. Phys.: Condens. Matter 2007, 19, 413101. (11) (a) Patra, D.; Pagliuca, C.; Subramani, C.; Samanta, B.; Agasti, S. S.; Zainalabdeen, N.; Caldwell, S. T.; Cooke, G.; Rotello, V. M. Chem. Commun. 2009, 4248−4250. (b) Skaff, H.; Lin, Y.; Tangirala, R.; Breitenkamp, K.; Boker, A.; Russell, T. P.; Emrick, T. Adv. Mater. 2005, 17, 2082−2086. (c) Arumugam, P.; Patra, D.; Samanta, B.; Agasti, S. S.; Subramani, C.; Rotello, V. M. J. Am. Chem. Soc. 2008, 130, 10046−10047. (12) (a) Patra, D.; Ozdemir, F.; Miranda, O. R.; Samanta, B.; Sanyal, A.; Rotello, V. M. Langmuir 2009, 25, 13852−13854. (b) B Wang, B.; Wang, M.; Zhang, H.; Sobal, N. S.; Tong, W.; Gao, C.; Wang, Y.; Giersig, M.; Wang, D.; Moehwald, H. Phys. Chem. Chem. Phys. 2007, 9, 6313−6318. (13) Miller, R.; Fainerman, V. B.; Makievski, A. V.; Kragel, J.; Grigoriev, D. O.; Kazakov, V. N.; Sinyachenko, O. V. Adv. Colloid Interface Sci. 2000, 86, 39−82. (14) Rosen, M. J. Surfactants and Interfacial Phenomena, 3rd ed.; Wiley-Interscience: Hoboken, NJ, 2004. (15) (a) Butt, H.-J.; Graf, K.; Kappl, M. Physics and Chemistry of Interfaces, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2006. 2027

dx.doi.org/10.1021/la204017z | Langmuir 2012, 28, 2023−2027